4-arylpiperdines have long been recognized as a valuable class of compounds capable of affecting the central nervous system in animals. Zimmerman, in U.S. Pat. No. 4,191,771 (1980) discloses a process for preparing 1,3,4-trisubstituted-4-arylpiperidines which are described as being analgesics and narcotic antagonists. Zimmerman in U.S. Pat. No. 4,284,635 (1981) discloses 1,2,4,5-tetraalkyl-4-arylpiperidines as being potent analgesics. The procedure disclosed by Zimmerman for preparing these 4-arylpiperidines involves a 1,3,5-trisubstituted-l,4,5,6-tetrahydropyridine intermediate. The preparation procedure disclosed in the U.S. Pat. No. 4,191,771 has the disadvantage of employing diazomethane which is both explosive and toxic.
It would be advantageous to have an improved process for the preparation of the compounds of the Zimmerman patents. It has now been found that the 1,3,4,4-tetrasubstituted-1,2,3,4-tetrahydropyridine intermediate to the compounds of Zimmerman can be conveniently prepared by a process which involves heating a 4-carbonate-1,3,4-substituted-piperidine to eliminate the carbonate moiety followed by a metalloenamine alkylation. The carbonate is preferably resolved prior to the elimination reaction. The elimination reaction provides the 1,3,4-substituted-1,2,3,6-tetrahydropyridine derivative which can be alkylated to provide the desired intermediate.
It has now been found that certain 4-carbonate derivatives provide the desired unsaturated compound in high yields. By contrast elimination of 4-hydroxy-4-arylpiperidines under acidic conditions have been found to primarily provide 1,2,5,6-tetrahydropyridines. Casy et al., Tetrahedron, 23, 1405-1410 (1967) reported that refluxing in aqueous HCl provided a 1.0:2.3 ratio of 5--CH.sub.3 to 3--CH.sub.3 olefin products. Barnett, et al., J. Org. Chem., 54, 4795-4800 (1989), reported a similar 1.0:2.3 ratio of 5--CH.sub.3 olefin to 3--CH.sub.3 olefin with toluenesulfonic acid in toluene at reflux; however, 100% H.sub.3 PO.sub.4 at 100.degree. C. provided a 1.0:9.0 ratio of the 5--CH.sub.3 olefin to 3--CH.sub.3 olefin.
The thermal elimination of esters is wellknown as reported by DePuy and King, "Pyrolytic Cis Eliminations", Chem. Rev. , 431-457 (1960). Such thermal eliminations in a piperidine ring have been reported by Larson, et al., J. Med. Chem., 1973, Vol. 16, 195-198. Diamond, et al., J. Med. Chem., 1964, Vol. 7, 57-60, disclose eliminations in a nitrogen-containing heterocycle with a basic nitrogen but not a piperidine. The thermal elimination of certain carbonates to produce olefins has also been reported in steroidial molecules, e.g., O'Conner, et al., J. Am. Chem. Soc., 75, 2118-2123 (1953). However the compounds disclosed in this reference do not contain a basic nitrogen. This is in contrast with the instant carbonates which contain a basic nitrogen which have been found in the instant invention to provide the desired olefin in high yields. Rahman has reported the preparation of certain 1,3-dimethyl-4-phenyl-4-piperidyl carbonates. Indian J. Chem., Vol. 16B, 632-633 (1978). However, there is no suggestion of heating the carbonate to provide the corresponding tetrahydropyridine compound.